» Mon Jul 12, 2010 6:05 am So I decided to investigate the issue further, and found a rather interesting article here: http://www.psy.unipd.it/~tressold/cmssimple/uploads/includes/MetaFreeResp010.pdf
Basically, the first part is a study which shows the evidence for psi.
The second part is a criticism.
The third part, is a rebuttal to that criticism.
From first part:
It may also be possible that the evidence obtained in the present
study has a strong bearing on the debate about much-discussed
declines in ganzfeld research, and the alleged unreliability of psi
effects. Although we found evidence of a weak decline across five
ganzfeld databases, we add that appearances can be deceiving:
There is good evidence that the decline is “in decline,” with effects
showing an upward trend (see Figures 2 and 3). Furthermore, after
we removed four outlier studies, there was a nonsignificant decline.
In addition, with our study, the two significant databases
(total N 45), and only one small nonsignificant database (N
14), as well as, more generally, only one dubious meta-anolysis out
of five (see Figure 3), the two negative assumptions made against
ganzfeld research (i.e., that psi effects are in decline and are
unreliable) are undermined. Whichever of our anolyses is preferred,
it appears that 34 years of ganzfeld research has more often
than not produced a communications anomaly worth investigating
further, as evidenced by the cumulative record: 74 (72.5%; N
102) had positive z scores, even though only 27 (26.5%) of the 102
studies were independently significant ( .05). If 26.5% seems
inconclusive or ambiguous, then in spite of, or due to, the statistical
evidence, parapsychologists may still have some way to go to
convince skeptics...
In closing, we emphasize how important it is to free up this
line of investigation from unwarranted skepticism and hasty
judgments, so that these communication anomalies might be
treated and investigated in like manner with other psychological
functions.
study has a strong bearing on the debate about much-discussed
declines in ganzfeld research, and the alleged unreliability of psi
effects. Although we found evidence of a weak decline across five
ganzfeld databases, we add that appearances can be deceiving:
There is good evidence that the decline is “in decline,” with effects
showing an upward trend (see Figures 2 and 3). Furthermore, after
we removed four outlier studies, there was a nonsignificant decline.
In addition, with our study, the two significant databases
(total N 45), and only one small nonsignificant database (N
14), as well as, more generally, only one dubious meta-anolysis out
of five (see Figure 3), the two negative assumptions made against
ganzfeld research (i.e., that psi effects are in decline and are
unreliable) are undermined. Whichever of our anolyses is preferred,
it appears that 34 years of ganzfeld research has more often
than not produced a communications anomaly worth investigating
further, as evidenced by the cumulative record: 74 (72.5%; N
102) had positive z scores, even though only 27 (26.5%) of the 102
studies were independently significant ( .05). If 26.5% seems
inconclusive or ambiguous, then in spite of, or due to, the statistical
evidence, parapsychologists may still have some way to go to
convince skeptics...
In closing, we emphasize how important it is to free up this
line of investigation from unwarranted skepticism and hasty
judgments, so that these communication anomalies might be
treated and investigated in like manner with other psychological
functions.
From second:
My first attempt to critique parapsychological research (Hyman,
1957) motivated me to scrutinize actual experimental reports
rather than general summaries. Such scrutiny takes time and effort.
I was surprised to discover that parapsychologists were more
statistically and methodologically sophisticated than critics had
portrayed them. However, I also found disturbing examples of
methodological oversights along with otherwise admirable controls.
Perhaps the most puzzling experience came from my hours
devoted to a detailed inspection of the experiments in the original
ganzfeld database (Hyman, 1985). Even today I find it difficult to
understand how parapsychologists could have tolerated so many
obvious flaws in what was claimed to be their most successful
database. I found the autoganzfeld experiments greatly improved
in methodology over the original ganzfeld experiments. On the
other hand, my careful anolysis of the actual data from these
experiments uncovered peculiar patterns that could possibly point
to some subtle biases (Hyman, 1994).
To me the most bothersome aspect of parapsychological research
during the century or so of its existence is its persistent
inconsistency. During the past 50 years, I have become acquainted
with many parapsychologists who agreed with my assessment.
They were understandably distressed by this state of affairs. I have
already referred to some contemporary parapsychologists who
acknowledge the elusiveness and inconsistency of parapsychological
evidence (Atmanspacher & Jahn, 2003; Bierman, 2001;
Kennedy, 2001, 2003; Lucadou, 2001).
1957) motivated me to scrutinize actual experimental reports
rather than general summaries. Such scrutiny takes time and effort.
I was surprised to discover that parapsychologists were more
statistically and methodologically sophisticated than critics had
portrayed them. However, I also found disturbing examples of
methodological oversights along with otherwise admirable controls.
Perhaps the most puzzling experience came from my hours
devoted to a detailed inspection of the experiments in the original
ganzfeld database (Hyman, 1985). Even today I find it difficult to
understand how parapsychologists could have tolerated so many
obvious flaws in what was claimed to be their most successful
database. I found the autoganzfeld experiments greatly improved
in methodology over the original ganzfeld experiments. On the
other hand, my careful anolysis of the actual data from these
experiments uncovered peculiar patterns that could possibly point
to some subtle biases (Hyman, 1994).
To me the most bothersome aspect of parapsychological research
during the century or so of its existence is its persistent
inconsistency. During the past 50 years, I have become acquainted
with many parapsychologists who agreed with my assessment.
They were understandably distressed by this state of affairs. I have
already referred to some contemporary parapsychologists who
acknowledge the elusiveness and inconsistency of parapsychological
evidence (Atmanspacher & Jahn, 2003; Bierman, 2001;
Kennedy, 2001, 2003; Lucadou, 2001).
From third:
Hyman (2010) is convinced of parapsychology’s “persistent
inconsistency” (p. 489), but we maintain that psi is anything but
inconsistent. We argue that the meta-anolytic results do count for
something. It is of paramount importance not to dismiss statistical
anomalies as nothing more than numerical oddities; we would
hope that on the strength of the peculiarity of the findings in and
of themselves, the scientifically inclined will be sufficiently intrigued
by the curious nature of psi to want to find out for
themselves what that anomaly might actually be.
On that point, Hyman (2010) claims the following:
"Parapsychology will succeed in its quest to demonstrate its communications
anomaly only when it can generate specific hypotheses that
predict patterns of outcomes that are consistent, lawful, and independently
replicable by parapsychologists and others. So far, careful
assessment of the parapsychological literature does not justify optimism
on this matter." (p. 490)
In response, we believe had addressed these issues, and argue for a
shift in mind-set: Instead of parapsychologists’ giving the null hypothesis
a chance (Alcock, 2003), skeptics should give the alternative
hypothesis a chance. In spite of the relatively limited pool of literature,
we argue that consistency has been demonstrated in the data and that
there is good evidence of replication by a range of investigators.
inconsistency” (p. 489), but we maintain that psi is anything but
inconsistent. We argue that the meta-anolytic results do count for
something. It is of paramount importance not to dismiss statistical
anomalies as nothing more than numerical oddities; we would
hope that on the strength of the peculiarity of the findings in and
of themselves, the scientifically inclined will be sufficiently intrigued
by the curious nature of psi to want to find out for
themselves what that anomaly might actually be.
On that point, Hyman (2010) claims the following:
"Parapsychology will succeed in its quest to demonstrate its communications
anomaly only when it can generate specific hypotheses that
predict patterns of outcomes that are consistent, lawful, and independently
replicable by parapsychologists and others. So far, careful
assessment of the parapsychological literature does not justify optimism
on this matter." (p. 490)
In response, we believe had addressed these issues, and argue for a
shift in mind-set: Instead of parapsychologists’ giving the null hypothesis
a chance (Alcock, 2003), skeptics should give the alternative
hypothesis a chance. In spite of the relatively limited pool of literature,
we argue that consistency has been demonstrated in the data and that
there is good evidence of replication by a range of investigators.
You guys are my favorite group of internet skeptics and believers, with a wide range of opinions, so I hope you find this interesting!
